Page 1

Newsletter of the Center for Proton Therapy  ::  Paul Scherrer Institut  ::  December 2017  ::  # 13 :: page 1


Center for Proton Therapy  ::  Paul Scherrer Institut  ::  # 13_12/2017

Dear Reader,

markable radiological (see figure) and clinical response observed van den Water, PhD. The investigators looked at a planning of a in a patient with advanced sacral chordoma treated both with hy-

sino-nasal malignancy with both a standard and a reduced number

it is my distinct pleasure to present you with our latest issue and perthermia (delivered at KSA, Aarau) and proton therapy. PSI and of spots. For the spot-reduced plan, the number of spots was minlast 2017 edition of SpotOn+. I would like to take the opportunity of KSA have engaged in a fruitful clinical collaboration and these two imized using the ‘pencil beam resampling’ technique as described this newsletter to inform you that after the elective shutdown of our

institutions are currently piloting a prospective protocol for non-re-

in the article. Interestingly, the number of spots could be reduced

cyclotron this summer we have resumed our full clinical patient sectabale sarcoma (STS). Proton therapy was delivered with a SIB from 26,069 in the clinical plan to 1,540 in the spot-reduced plan, throughput. In this edition, we report the assessment of the preva-

paradigm. We have treated four additional patients and the (very) a roughly 95% decrease with no plan corruption. Moreover, the

lence rate of brain necrosis in children treated for a brain tumor at initial results seem promising. As such, we are considering launch-

spot-reduced plan was found to be clinically deliverable and impor-

PSI. As you know, the brain necrosis-issue has been looming in the ing a clinical protocol with possibly the active international collab-

tantly the delivery time per field was shortened on average from 56

radiation Oncology for quite some time. We decided late 2016 to

oration with the Sarcoma Unit of University College London Hospital, s to 20 s (i.e. 65% reduction in delivery time). Interestingly, the

analyze our results in children treated with PBS protons. This is a

UK. One clinical observation of paramount importance during the

spot-re¬duced plan was less sensitive to rigid setup and range errors,

large cohort (170+ patients) composed of young to very young pa-

treatment of these challenging patients was the importance of pain

with a target mean dose reduction of < 0.5% in the worst case sce-

tients (median age, 3.3 years) treated with PBS protons. This joint control during and immediately after the combined therapeutic nario, compared to a > 2.0% reduction for the standard spot plan. Inselspital/PSI assessment was made scrupulously with a neuro-ra-

modalities. Last but not least, treatment time is of critical importance

These results are interesting but should be confirmed by other

diologist looking for brain necrosis (BN) and/or white matter

in photon and proton therapy alike. Shortening radiation delivery simulations. Definitively, delivering non-standard spot plans may

changers, regardless if the patient was symptomatic or not. Roughly

times, by reducing the number of proton spots (i.e. pencil beams)

be clinically advisable. With this last controversial report, I would

60% of BN were asymptomatic and identified only on routine MRI while maintaining dosimetric plan quality may present a number of like to take the opportunity to wish you all a very merry ‘Xmas and scans. All save one were observed within a 2 years period. Interest-

advantage, including but not limited to patient comfort, intra-frac-

happy new year for 2018.

ingly, ependymoma histology was a risk factor, among others, for tional patient motion, shorter anesthesia duration for sedated this radiation-induced complication. It is fair to say, that we observed children and patient throughput to name a few. In this issue, Franc-

Yours sincerely,

a low prevalence of symptomatic BN and these data are in line with esca Belosi reports the results of a joint study performed with the

Prof. Damien Charles Weber

other modern photon series. The second article describes the re-

Erasmus MC Cancer Institute, Rotterdam, Netherlands by Steven

Chairman of CPT, Paul Scherrer Institute

Newsletter of the Center for Proton Therapy  ::  Paul Scherrer Institut  ::  December 2017  ::  # 13 :: page 2

Radio-Oncology News Pediatric Patients with Brain Tumors treated with Pencil Beam Scanning Proton Therapy: complications of brain necrosis and whiter matter lesions Introduction

ation-induced inflammation, causing

Materials and Methods

progenitor cells to differentiate into Successful treatment of brain tumors glial cells instead of neurons. Although Between 1999 and 2015, 171 pediatric in children often requires an interdis-

histopathology can be used to estab-

patients (<18 years) were treated with

ciplinary strategy. In addition to sur-

lish a diagnosis of RN or WML, more

PBS PT. Median age at diagnosis was

gery, chemotherapy and radiotherapy commonly a clinicoradiologic diagnosis 3.3 years (range, 0.3-17.0) and the meare essential pillars in the overall con-

of radiation necrosis is used to avoid dian delivered dose was 54 Gy (RBE)

cept. Compared with conventional surgical morbidity and potential com-

(range, 40–74.1). Radiation necrosis

(photon) radiotherapy, proton therapy plications. Some authors have reported and WML were defined as a new area (PT) offers the physical advantages of a very low prevalence of RN (0.6 %) a reduced entrance dose, the absence

of abnormal signal intensity on T2- 2 patients, respectively. Eighteen (11%) photon radiation therapy. Pre-PT

after PT, whilst others have reported a weighted images or increased signal patients developed WML at a median

of an exit dose and thereby provides a high rate (43%) of this feared compli-

chemotherapy or any chemotherapy

intensity on T2-weighted images, and time of 14.5 months (range, 2-62), most administration, ependymomal tumors

highly conformal dose distribution in

cation. This might be related to mostly contrast enhancement on T1 occurring of them (n=13; 72%) being asympto-

and hydrocephalus as an initial symp-

the target volume, and a reduced radi-

non-specific imaging techniques or in the brain parenchyma included in

tom were significant risk factors asso-

ation dose to adjacent healthy tissue. differences in clinical terminology leadThese advantages of protons over pho-

the radiation treatment field, which did toxicities occurred in 4 and 1 patients,

ing to varying estimates of toxicities not demonstrate any abnormality be-

tons led to an increased use of PT in caused by treatment induced brain

matic (grade 1). WML Grade 2 and 3

respectively. The 5-year RN-free and This evaluation was done in a cooper-

fore PT. Radiation necrosis and WML WML-free survival was 83% and 87%, ation between Inselspital Bern and PSI

pediatric brain tumors. Parenchymal lesions. However, limited data exist were graded according to the Common respectively. In univariate analysis, brain alterations such as radiation ne-

regarding RN or WMLs after PT in pedi-

ciated with RN.

Terminology Criteria for Adverse

by a resident staying one year at PSI.

neo-adjuvant (p=0.025) or any The results were presented this year at

crosis (RN) or white matter lesions atric patients. Further investigation of Events. The median follow-up period (p=0.03) chemotherapy, hydrocepha-

the SIOP-conference in Washington

(WML) are feared complications occur-

and will be published soon.

MRI-parenchymal brain alterations in

ring after radiotherapy. Usually this is pediatric patients with a brain tumor is

for the surviving patients was 49.8

lus before PT (p=0.035), and ependy-

months (range, 5.9-194.7).

moma (p=0.026) histology were signif-

treated with steroids, but more recent required to understand the incidence,

icant predictors of RN.

data suggests that anti-VEGF agents timing, clinical significance and risk Results such as bevacizumab may also be ef- factors of RN and WMLs in children


Reference Bojaxhiu et al. Radiation Necrosis

fective since it was established that RN

treated with PT. As such, the purpose Twenty-nine (17%) patients developed

might be mediated either by an is-

of this study was to assess the rate of RN at a median time of 5 months Children treated with PT demonstrated ric Patients with Brain Tumors

and White Matter Lesions in Pediat-

chemic response or oligodendrocyte RN and WMLs in a cohort of brain tumor

(range, 1-26), most of them (n=17; 59%) a low prevalence of symptomatic RN

treated with Pencil Beam Scanning

damage, while white matter lesions children treated with pencil beam scan-

being asymptomatic (grade 1). Grade

Proton Therapy. Int J Rad Onc Biol

(WML) might be the result of the radi-

2, 4 and 5 toxicities occurred in 8, 2 and cohorts treated with either proton or Phys (accepted for publication)

ning (PBS) PT.

(7%) or WML (3%) compared to similar

Newsletter of the Center for Proton Therapy  ::  Paul Scherrer Institut  ::  December 2017  ::  # 13 :: page 3

Radio-Oncology News Combination of proton therapy and hyperthermia in selected tumor types show promising results The combination of hyperthermia with

comas arise often in the soft tissue of ing organs at risk. The synergy of proton

radiation therapy has led to higher re-

the extremities or the trunk. A resection therapy with hyperthermia leads to

sponse rates in different tumor types would lead either to an amputation or maximal tumor-cytotoxic effect in the as sarcomas, cervical cancer, breast a very mutilating disfigurement and cancer. Hyperthermia for deep seated cancer, rectal cancer, melanomas or

often it is not possible to remove the

bladder cancer. In Switzerland there

whole tumor especially in the area of electromagnetic waves in form of mi-

tumors is generated by the exposure to

are only few institutions offering hyper-

the trunk. The alternative treatment is crowaves. It is applied via regional

thermia therapy, the largest of them is radiation therapy. But the applicable

deep hyperthermia and heats up the

the radiation oncology department of dose with conventional radiation ther-

tumor tissue to temperatures in the

Hyperthermia unit in the radiation oncology department at Kantonsspital Aarau. Preparation of a treatment session by the team (left to right): N.Lomax, O. Timm, Dr. E. Puric

Kantonsspital Aarau (head: Prof.

apy is often limited due to the delicate range of 41.5-42.5°C. Hyperthermia at proton therapy. Up to now 5 patients slight regression or stable tumor situ-

Stephan Bodis).

localisations of the tumors surrounded these temperatures enhances the radi-

The Center for Proton Therapy CPT at by sensitive healthy organs. The bene-

were treated within a preliminary fea-

ation. Initial clinical pretreatment

ation induced damage in the tumor cell. sibility treatment protocol. Proton ther-

symptoms showed clear decrease dur-

PSI has a close collaboration with KSA,

ficial physical behaviors of protons This thermal sensitizing effects in-

apy was delivered in a simultaneous ing follow up in all patients. The imple-

among others in the framework of a

compared with photons that are used crease on one hand the sensitivity of boost treatment concept, meaning that mentation of a clinical trial protocol is

prospective phase I/II clinical trial (HY-

in conventional radiation therapy allow hypoxic, nutritionally deficient tumor

PROSAR-study, NCT01904565) on the

delivering higher doses in the tumor

cells in low pH, on the other hand it a higher total dose than the surround-

These early results on sacral chordo-

treatment of soft tissue sarcomas. Sar-

area without overdosing the surround-

inhibits the radiation induced DNA

ing area with microscopic spread in the

mas have been submitted to the annual

damage repair capacity of same number of treatment sessions.

meeting 2018 of the European Society

the tumor cells.

for Radiotherapy&Oncology (ESTRO)

the macroscopic gross tumor received currently under discussion.

This leads to a hypofractionated irradi-

Based on the experience ation of the gross tumor with higher next spring in Barcelona, Spain. with the HYPROSAR-study single fraction dose, which especially the use of proton therapy in radio-resistant tumors show addi-

For any information,

with hyperthermia was tional positive effect. With this concept please refer to, introduced in clinical the treatment lasts 5 ½ weeks with 5 Dr. Marc Walser, CPT

MRI’s of the pelvis of a 73 year old patient with a sacral chordoma. MRI before (left image) and 9 months after combined hyperthermia and proton therapy (right image) showing a clear decrease of tumor size.

practice for selected large

treatment sessions per week. Hyper-

Tel. +41 56 310 59 48

inoperable sacral chordo-

thermia is given once a week before or

mas which usually show

after proton therapy. Early results in

unsatisfactory local tumor

these patients are encouraging with no

control rates even when

higher grade acute and late toxicities. Tel. +41 62 838 95 59

treated with high dose

Early follow-up imaging showed either

Prof. Dr. N.R. Datta, KSA

Newsletter of the Center for Proton Therapy  ::  Paul Scherrer Institut  ::  December 2017  ::  # 13 :: page 4

Medical-Physics News Reduced spot numbers for PBS proton therapy shortens delivery time without dosimetric plan compromise. Purpose

For both plans, the same 4-beams arrangement was ard deviation of used with 4mm lateral spot spacing, 2.5mm energy the linear correla-

The degeneracy of spot-scanned proton therapy layer spacing and a 4.2cm water-equivalent pre-ab-




treatment plans can be exploited to shorten delivery sorber. The planning target volume was 280 cm3

spot-reduced plan

times, by reducing the number of proton spots (i.e. and received a homogeneous dose from each field. and <2% (standard pencil beams) while maintaining dosimetric plan

Figure 1: Planned dose distributions and log-file recalculations (delivered-planned) for the total dose and the dose of field 1 for the clinical plan (left hand-side) and the spot-reduced plan (right-hand side).

For the spot-reduced plan, the number of spots was

deviation) for the

quality. Because a strongly reduced number of spots minimized using the ‘pencil beam resampling’ tech-

clinical plan. For

can potentially affect the treatment delivery, we

nique, which involves repeated inverse optimiza-

both plans, the

performed an extensive investigation to assess the

tion, while adding in each iteration randomly se-

log-file recalculated dose was within ±1% of the tial delivery time reduc-

deliverability, delivery accuracy, robustness and lected spots and excluding low-weighted spots planned dose for 100% of the voxels (96% and 98% tion of 65% on average per actual delivery time reduction of a spot-reduced until the plan quality deteriorates. Machine steering on average for the individual fields of the spot-re-


treatment plan.

spot-reduced plan was

files were generated and both treatment plans were duced or clinical plan, respectively) (Figure 1). The



delivered on our PBS Gantry 2 at PSI, comparing the

robustness simulations showed that random spot more sensitive to delivery

Material and methods

delivery time per field, measured dose profiles in

position errors of ≤0.5mm resulted in 94%/100% of uncertainties, the accu-

water and recalculated dose distributions using

voxels passing the ±1% criterion for the spot-re-

racy of total delivered

For one patient treated in the sinonasal region, a

log-files. In addition, simulations were performed duced/clinical plans, respectively. Surprisingly, the

dose was well within clin-

‘spot-reduced plan’ was generated using Eras-

to compare the robustness against random errors spot-reduced plan was less sensitive to rigid setup

ical tolerance.

mus-iCycle (Erasmus MC Cancer Institute, Rotter-

in individual spot position and against systematic and range errors, with a target mean dose reduction

This work has been sub-

dam, Netherlands) for which the dosimetric plan

errors in patient setup (±3mm along 3 axes) and of 0.4% in the worst case scenario, compared to a

mitted to the annual meet-

quality was equal or better than the clinical plan proton range (±3%).

2.4% reduction for the standard plan. The increased ing 2018 of the European

generated using PSIplan (PSI, Villigen, Switzerland).

dose to organs at risk was of 2.6%/20.4% as max-

Society for Radiotherapy-

imum dose to the brainstem in the worst case sce-

&Oncology (ESTRO) next

nario for the spot-reduced and standard plan, re-

spring in Barcelona,


The number of spots was reduced by 94% from spectively.


Imprint Editor Dr. Ulrike Kliebsch Chairman Prof. Damien C. Weber Chief Medical Physicist Prof. Tony Lomax Design and Layout Monika Blétry


measured dose profiles showed differences between

be reduced by 94%, without compromising the

Tel. +41 56 310 37 45,

Center for Proton Therapy CH-5232 Villigen PSI Tel. +41 56 310 35 24 Fax +41 56 310 35 15

delivered and planned dose of 2.9%-4.3% (as stand-

dosimetric plan quality, which resulted in a substan-

Villigen PSI, December 2017

26069 in the clinical plan to 1540 in the spot-reduced plan. The spot-reduced plan was found to be deliverable and the delivery time per field was shortened


For any further information, please refer to CPT,

by 65% on average from 56 s to 20 s (Table 1). The Compared with the clinical plan, spot number could Maria Francesca Belosi

PSI Protontherapy Newsletter Nr. 13 (12/2017)  
PSI Protontherapy Newsletter Nr. 13 (12/2017)